PROBE recasts MLIP uncertainty quantification as selective classification by training a compact discriminative classifier on frozen per-atom backbone embeddings, yielding a reliability probability that tracks actual error better than ensemble disagreement.
Title resolution pending
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
representative citing papers
A single autoregressive foundation model unifies protein, molecule, and crystal structures into a shared token vocabulary and generates inspectable reasoning traces, achieving SOTA on 67 of 86 scientific tasks.
ERA combines LLMs and tree search to produce expert-level empirical software that outperforms top human methods on single-cell analysis leaderboards and CDC COVID-19 forecasts.
A hierarchical ALIGNN + ALIGNN-FF + DFT pipeline recovers experimental average voltages within 0.3 V on four commercial Li-ion cathodes and ranks large materials databases into surrogate-level shortlists.
A generative solver separates data-driven prior learning from inference-time enforcement of conservation laws using martingale-regularized score matching and physics-informed sampling for stable field reconstruction.
citing papers explorer
-
Knowing when to trust machine-learned interatomic potentials
PROBE recasts MLIP uncertainty quantification as selective classification by training a compact discriminative classifier on frozen per-atom backbone embeddings, yielding a reliability probability that tracks actual error better than ensemble disagreement.
-
Accurate, Interdisciplinary and Transparent Structure-property Understanding with Deep Native Structural Reasoning
A single autoregressive foundation model unifies protein, molecule, and crystal structures into a shared token vocabulary and generates inspectable reasoning traces, achieving SOTA on 67 of 86 scientific tasks.
-
An AI system to help scientists write expert-level empirical software
ERA combines LLMs and tree search to produce expert-level empirical software that outperforms top human methods on single-cell analysis leaderboards and CDC COVID-19 forecasts.
-
BatteryMat: a hierarchical machine-learning and DFT framework for average-voltage screening of lithium-ion cathode materials
A hierarchical ALIGNN + ALIGNN-FF + DFT pipeline recovers experimental average voltages within 0.3 V on four commercial Li-ion cathodes and ranks large materials databases into surrogate-level shortlists.
-
Physics-Informed Generative Solver: Bridging Data-Driven Priors and Conservation Laws for Stable Spatiotemporal Field Reconstruction
A generative solver separates data-driven prior learning from inference-time enforcement of conservation laws using martingale-regularized score matching and physics-informed sampling for stable field reconstruction.